Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Mon, 12 Oct 2020 14:20:41 +0200
From:   Maxime Ripard <maxime@cerno.tech>
To:     Samuel Holland <samuel@sholland.org>
Cc:     =?utf-8?B?Q2zDqW1lbnQgUMOpcm9u?= <peron.clem@gmail.com>,
        Chen-Yu Tsai <wens@csie.org>, Rob Herring <robh+dt@kernel.org>,
        Mark Brown <broonie@kernel.org>,
        Liam Girdwood <lgirdwood@gmail.com>,
        Jaroslav Kysela <perex@perex.cz>,
        Takashi Iwai <tiwai@suse.com>,
        Marcus Cooper <codekipper@gmail.com>,
        Jernej Skrabec <jernej.skrabec@siol.net>,
        alsa-devel@alsa-project.org, devicetree@vger.kernel.org,
        linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org,
        linux-sunxi@googlegroups.com
Subject: Re: [PATCH v6 02/14] ASoC: sun4i-i2s: Change set_chan_cfg() params
Message-ID: <20201012122041.nf6vrozeq3zq2kqj@gilmour.lan>
References: <20201003141950.455829-1-peron.clem@gmail.com>
 <20201003141950.455829-3-peron.clem@gmail.com>
 <20201005121307.v6jpyeyfi4kxc2cl@gilmour.lan>
 <fae04296-9865-3dcb-e02d-46062bf778a8@sholland.org>
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha256;
        protocol="application/pgp-signature"; boundary="z7kb5v5tpj7u77hs"
Content-Disposition: inline
In-Reply-To: <fae04296-9865-3dcb-e02d-46062bf778a8@sholland.org>
Precedence: bulk


--z7kb5v5tpj7u77hs
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Tue, Oct 06, 2020 at 12:03:14AM -0500, Samuel Holland wrote:
> On 10/5/20 7:13 AM, Maxime Ripard wrote:
> > On Sat, Oct 03, 2020 at 04:19:38PM +0200, Cl=E9ment P=E9ron wrote:
> >> As slots and slot_width can be set manually using set_tdm().
> >> These values are then kept in sun4i_i2s struct.
> >> So we need to check if these values are setted or not
> >> in the struct.
> >>
> >> Avoid to check for this logic in set_chan_cfg(). This will
> >> duplicate the same check instead pass the required values
> >> as params to set_chan_cfg().
> >>
> >> This will also avoid a bug when we will enable 20/24bits support,
> >> i2s->slot_width is not actually used in the lrck_period computation.
> >>
> >> Suggested-by: Samuel Holland <samuel@sholland.org>
> >> Signed-off-by: Cl=E9ment P=E9ron <peron.clem@gmail.com>
> >> ---
> >>  sound/soc/sunxi/sun4i-i2s.c | 36 ++++++++++++++----------------------
> >>  1 file changed, 14 insertions(+), 22 deletions(-)
> >>
> >> diff --git a/sound/soc/sunxi/sun4i-i2s.c b/sound/soc/sunxi/sun4i-i2s.c
> >> index c5ccd423e6d3..1f577dbc20a6 100644
> >> --- a/sound/soc/sunxi/sun4i-i2s.c
> >> +++ b/sound/soc/sunxi/sun4i-i2s.c
> >> @@ -177,8 +177,9 @@ struct sun4i_i2s_quirks {
> >>  	unsigned long (*get_bclk_parent_rate)(const struct sun4i_i2s *);
> >>  	s8	(*get_sr)(const struct sun4i_i2s *, int);
> >>  	s8	(*get_wss)(const struct sun4i_i2s *, int);
> >> -	int	(*set_chan_cfg)(const struct sun4i_i2s *,
> >> -				const struct snd_pcm_hw_params *);
> >> +	int	(*set_chan_cfg)(const struct sun4i_i2s *i2s,
> >> +				unsigned int channels,	unsigned int slots,
> >> +				unsigned int slot_width);
> >>  	int	(*set_fmt)(const struct sun4i_i2s *, unsigned int);
> >>  };
> >> =20
> >> @@ -414,10 +415,9 @@ static s8 sun8i_i2s_get_sr_wss(const struct sun4i=
_i2s *i2s, int width)
> >>  }
> >> =20
> >>  static int sun4i_i2s_set_chan_cfg(const struct sun4i_i2s *i2s,
> >> -				  const struct snd_pcm_hw_params *params)
> >> +				  unsigned int channels, unsigned int slots,
> >> +				  unsigned int slot_width)
> >>  {
> >> -	unsigned int channels =3D params_channels(params);
> >> -
> >>  	/* Map the channels for playback and capture */
> >>  	regmap_write(i2s->regmap, SUN4I_I2S_TX_CHAN_MAP_REG, 0x76543210);
> >>  	regmap_write(i2s->regmap, SUN4I_I2S_RX_CHAN_MAP_REG, 0x00003210);
> >> @@ -434,15 +434,11 @@ static int sun4i_i2s_set_chan_cfg(const struct s=
un4i_i2s *i2s,
> >>  }
> >> =20
> >>  static int sun8i_i2s_set_chan_cfg(const struct sun4i_i2s *i2s,
> >> -				  const struct snd_pcm_hw_params *params)
> >> +				  unsigned int channels, unsigned int slots,
> >> +				  unsigned int slot_width)
> >>  {
> >> -	unsigned int channels =3D params_channels(params);
> >> -	unsigned int slots =3D channels;
> >>  	unsigned int lrck_period;
> >> =20
> >> -	if (i2s->slots)
> >> -		slots =3D i2s->slots;
> >> -
> >>  	/* Map the channels for playback and capture */
> >>  	regmap_write(i2s->regmap, SUN8I_I2S_TX_CHAN_MAP_REG, 0x76543210);
> >>  	regmap_write(i2s->regmap, SUN8I_I2S_RX_CHAN_MAP_REG, 0x76543210);
> >> @@ -467,11 +463,11 @@ static int sun8i_i2s_set_chan_cfg(const struct s=
un4i_i2s *i2s,
> >>  	case SND_SOC_DAIFMT_DSP_B:
> >>  	case SND_SOC_DAIFMT_LEFT_J:
> >>  	case SND_SOC_DAIFMT_RIGHT_J:
> >> -		lrck_period =3D params_physical_width(params) * slots;
> >> +		lrck_period =3D slot_width * slots;
> >>  		break;
> >> =20
> >>  	case SND_SOC_DAIFMT_I2S:
> >> -		lrck_period =3D params_physical_width(params);
> >> +		lrck_period =3D slot_width;
> >>  		break;
> >> =20
> >>  	default:
> >> @@ -490,15 +486,11 @@ static int sun8i_i2s_set_chan_cfg(const struct s=
un4i_i2s *i2s,
> >>  }
> >> =20
> >>  static int sun50i_h6_i2s_set_chan_cfg(const struct sun4i_i2s *i2s,
> >> -				      const struct snd_pcm_hw_params *params)
> >> +				      unsigned int channels, unsigned int slots,
> >> +				      unsigned int slot_width)
> >>  {
> >> -	unsigned int channels =3D params_channels(params);
> >> -	unsigned int slots =3D channels;
> >>  	unsigned int lrck_period;
> >> =20
> >> -	if (i2s->slots)
> >> -		slots =3D i2s->slots;
> >> -
> >>  	/* Map the channels for playback and capture */
> >>  	regmap_write(i2s->regmap, SUN50I_H6_I2S_TX_CHAN_MAP0_REG, 0xFEDCBA98=
);
> >>  	regmap_write(i2s->regmap, SUN50I_H6_I2S_TX_CHAN_MAP1_REG, 0x76543210=
);
> >> @@ -525,11 +517,11 @@ static int sun50i_h6_i2s_set_chan_cfg(const stru=
ct sun4i_i2s *i2s,
> >>  	case SND_SOC_DAIFMT_DSP_B:
> >>  	case SND_SOC_DAIFMT_LEFT_J:
> >>  	case SND_SOC_DAIFMT_RIGHT_J:
> >> -		lrck_period =3D params_physical_width(params) * slots;
> >> +		lrck_period =3D slot_width * slots;
> >>  		break;
> >> =20
> >>  	case SND_SOC_DAIFMT_I2S:
> >> -		lrck_period =3D params_physical_width(params);
> >> +		lrck_period =3D slot_width;
> >>  		break;
> >> =20
> >>  	default:
> >> @@ -565,7 +557,7 @@ static int sun4i_i2s_hw_params(struct snd_pcm_subs=
tream *substream,
> >>  	if (i2s->slot_width)
> >>  		slot_width =3D i2s->slot_width;
> >> =20
> >> -	ret =3D i2s->variant->set_chan_cfg(i2s, params);
> >> +	ret =3D i2s->variant->set_chan_cfg(i2s, channels, slots, slot_width);
> >=20
> > Isn't slots and slot_width set to 0 here ?
> >=20
> > And therefore, wouldn't we set lrck_period to 0 if we're using any
> > format but I2S?
> >=20
> > More importantly, I'm not really convinced this needs to be done, and it
> > introduces some side effects that are not explained.
>=20
> If I set dai-tdm-slot-width =3D <32> and start a stream using S16_LE,
> currently we would calculate BCLK for 32-bit slots, but program
> lrck_period for 16-bit slots, making the sample rate double what we
> expected. That sounds like a bug to me. (Because of that, as Chen-Yu
> mentioned in reply to v5, this should be the first patch in the series.)

I'm not denying that there's a bug though here :)

You've spent way more time than I did with this driver recently, so I
definitely take your word for it.

> Could you be more specific what side effects you are referring to?

I don't really like the change of semantics associated to the new
prototype, and it becomes less obvious what we're supposed to do with
slots and slot_width. Like, those are very TDM-y terms, are we supposed
to honour them when running in I2S, or should we just ignore them?

Maxime

--z7kb5v5tpj7u77hs
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----

iHUEABYIAB0WIQRcEzekXsqa64kGDp7j7w1vZxhRxQUCX4RKGAAKCRDj7w1vZxhR
xYi5AP4kX0Q+dHGX2Cx6jXMcXiMMTf3IKxRJD6Ia04QH56FSuwEA6XFXC0J1Wf0t
SdBHH14nRJv9tZbVyMLU+N67wO2mDAM=
=e1V4
-----END PGP SIGNATURE-----

--z7kb5v5tpj7u77hs--